GOA-ACO: A goose optimized ant colony algorithm for the automated guided vehicle path planning

Abstract

This article proposes a goose optimized ant colony algorithm (GOA-ACO) to enhance the quality and efficiency of path optimization for Automated Guided Vehicle(AGV) in intelligent production environments. By integrating initialized parameters, the fixed parameters of the ant colony optimization algorithm(ACO) are replaced with a dynamic adjustment mechanism optimized via the goose optimization algorithm (GOA), while a sound propagation model is introduced to construct a hybrid initial solution space. In each iteration, the pheromone utilization coefficient and heuristic weight of the ACO are adjusted through the single leg balance strategy(SLBS) optimized by the GOA, thereby achieving dynamic parameter collaborative updating. An embedded inspired wake-up mechanism is incorporated into the ant path search process, and random perturbation strategy(RPS) are added to the probability formula to enhance the diversity of path selection probabilities. Furthermore, the dimension scaling factor of the GOA is employed to dynamically compress the search space, which improves the optimization efficiency and convergence speed for high-dimensional problems. Experimental results indicate that the proposed GOA-ACO achieves significant improvements in path planning performance compared to benchmark algorithms, while demonstrating stronger adaptability to practical AGV operating environments.